AU2002214051B2

AU2002214051B2 - System for testing an electric high frequency signal and level measuring device provided with said system

Info

Publication number: AU2002214051B2
Application number: AU2002214051A
Authority: AU
Inventors: Josef Fehrenbach; Karl Griessbaum; Jurgen Motzer
Original assignee: Vega Grieshaber KG
Current assignee: Vega Grieshaber KG
Priority date: 2000-11-23
Filing date: 2001-11-13
Publication date: 2006-12-14
Anticipated expiration: 2021-11-13
Also published as: EP1336119A1; WO2002042793A1; EP1336119B9; DE10058026A1; AU1405102A; ATE509283T1; EP1336119B1; CN100347565C; CN1498347A

Abstract

The invention relates to a system for testing an electric high frequency signal, comprising at least one conductor element (6) wherein the electric high frequency signal is fed thereto at an input point (10) and transmitted at an output point (17) to a probe (7) which is used to guide the high frequency signal. The invention also comprises a single or multi-part mechanical support element (1), and a single or multi-part insulation (11,12) which is arranged between the support element (1) and the conductor element (6). The inventive system is characterised in that the impedance of said system and the impedance of the probe (7) connected to the output point (17) are substantially equalised at said output point (17).

Description

VEGA GRIESHABER KG WO 02/42793 Feedthrough for an electric radio frequency signal and filling level measuring device having such feedthrough Field of the invention The present invention relates to a feedthrough for an electric radio frequency signal such as it is generated for example in a filling level measuring device and evaluated upon reflection from a surface of a filling matter to be monitored. A feedthrough of this kind includes a conductor element into which at an inlet position the electric radio frequency signal is to be fed in and which, at an outlet position, transmits the radio frequency signal to a probe.

Further, the feedthrough comprises a mechanical carrier element having one or more parts.

Between the carrier element and the guiding element, there is an insulation having one or more parts. The invention further relates to filling level measuring means functioning according to the principle of the echo time measurement of guided electromagnetic waves and being provided with a feedthrough of the above mentioned type.

Description of the prior art For filling level measurement, measuring systems are used determining the distance to the filling matter based on the measured echo time of electromagnetic waves from a filling level measuring device mounted at the vessel top wall to the filling matter surface, and back. Due to a knowledge of the vessel's height, the required filling level height can be calculated. Such sensors, also known to those skilled in the art as "filling level radar", are all based on the property of electromagnetic waves to propagate at a constant velocity within a homogeneous non conductive medium and to be at least partially reflected at the interface of different UP:ar -2media. Each interface of the two media having different dielectric constants causes a radar echo to be generated upon the incidence of a wave. The greater the difference of the two dielectric constants the stronger the change in the wave resistance of the wave propagation and the greater the echo to be observed.

For determining the desired wave echo time various radar principles are known. The two main methods used are on the one hand the pulse echo time method (pulse radar) and on the other hand frequency modulated constant wave (FMCW) radar. Pulse radar uses pulse-shaped amplitude modulation of the wave to be radiated and determines the direct time delay between transmitting and receiving said pulses. The FMCW radar determines the echo time in an indirect way by transmitting a frequency modulated signal and forming a difference between a transmitted and a received momentary frequency.

Apart from the various radar principles, depending on the application, various frequency ranges of the electromagnetic waves are used. There are for example pulse radars having carrier frequencies in the range between 5 and 30 GHz and also those working in the base band as so-called mono pulse radars without carrier frequency.

Moreover, a series of methods and apparatuses are known to guide the electromagnetic wave to the filling matter surface and back. Herein, a principal distinction is made between a wave radiated into the space and a wave guided through a conductor. A filling level measuring device in which microwaves are fed through a coaxial cable into an antenna serving to radiate an electromagnetic wave is known from EP 0 834 722 A2. Here, the antenna is formed to have two parts. One antenna part in the form of a solid cylinder is of a dielectric material and surrounded by a metal sleeve. At the one end of the solid cylinder of dielectric material, the microwave is fed in, at the other end, the microwave is passed on to the radiating end of the antenna. The metal sleeve extends over the antenna area formed as the solid cylinder and situated in the area of a nozzle of a vessel containing the filling matter within it. This antenna structure, in particular the construction within the nozzle of the vessel, thus forms a filled hollow conductor for conducting the radio frequency signal or the wave to the area of the antenna serving for radiation. This structure has the effect that the antenna does not transmit microwaves in the area of the measuring device attachment i.e. the antenna portion lying in the area of the nozzle or receive reflected microwaves in this area of measuring device attachment. In order to avoid an impedance jump at the end of the metal sleeve facing the radiating antenna, the end of the sleeve is chamfered.

From EP 0 922 942 Al, a filling level measuring device working with microwaves and having a radiating antenna is also known. Herein the microwave fed in via a coaxial cable is fed into an end element formed with a taper at the side of the antenna. Adjacent thereto is an insert of a dielectric material having a recess of the end element corresponding to the taper. From this insert of a dielectric material, the microwave is passed on to the radiating antenna portions.

For achieving a quasi-continuous transition without a considerable impedance jump, the insert has a higher proportion of ceramic in the direction facing away from the antenna than in a section disposed in the transmitting direction facing the antenna.

Radar sensors having a completely different structure with respect to the feedthrough and signal guiding within the vessel and guiding the electromagnetic wave through a line (probe) to the location of reflection and back, are also called TDR (time domain reflectometry) sensors. These sensors have considerably lower loss of the reflected echo signal as compared with those which freely radiate radio frequency waves, since the power flow is only in a closely limited area in the ambient along the conductive wave guide. Also, interfering echoes from within the vessel, such as caused by reflections of the waves from vessel interior structures (stirring apparatus, conduits) and which in freely radiating sensors impede the identification of the one echo from the filling matter surface, are mostly avoided with sensors having guided waves. This results in the filling level measurement using guided electromagnetic waves being largely independent of the vessel construction and moreover of the product characteristics of the filling matter or other operating conditions dust, filling angle of bulk material) and therefore results in very reliable measuring results.

As a wave guide for guiding the wave, all the guides conductors normally used for radio frequency conduction may be used, in which the wave at least partially penetrates the medium surrounding, or being surrounded by, the metallic conductor. By its simple mechanical construction and its suitability for any filling matter, i.e. bulk material and liquids, in particular the one-wire conductor of one-conductor probe is often used in the filling level measurement industry. In its form as a rod or rope probe, it is largely unaffected by deposits and adhesions of filling matter. An exemplary filling level sensor having such a probe is described in DE 44 04 745 C2.

An important aspect of the TDR filling level sensors having single conductors is the coupling of the measuring signal from the electronics into the probe. It is essential that the conduction path from the electronics (electronics unit) to the probe does not have any major impedance jumps for the wave being conducted. For, at each jump change in the conduction impedance, part of the wave is reflected. This reflected portion on the one hand is no longer available for measuring purposes, i.e. reflection on the filling matter surface, whereby part of the amplitude of the echo generated there is lost. On the other hand, additional interfering echoes are generated by reflection of the wave at possible conduction impedance changes between the electronics and the probe, impeding the identification of the filling matter reflection to be evaluated. For in particular the interfering echo at the interface between the vessel feedthrough and the probe, depending on the band width of the measuring signal, extends over a distance area directly adjacent to the interface. With small echoes from the filling matter surface to be measured and a strong interference echo from the impedance jump at the beginning of the probe, it becomes impossible to detect and to precisely measure filling levels approaching the upper end of the probe. Thus with all known TDR sensors having a single conductor probe, there has to be a minimum distance between the filling matter and the signal feedthrough passing through the vessel wall which should not be undercut. It is usually about cm.

In filling level sensors, the conduction path between the electronics and the probe is always comprised of the abovementioned feedthrough and often also a coaxial cable communicating with the printed circuit board on which the electronic circuit for generating the transmission signal and for evaluating the reflected signal is constructed. The coaxial cable can be eliminated in some cases where the printed circuit board has a direct electrical and mechanical connection to the feedthrough.

The feedthrough has the function to guide the measuring signal from the sensor attached outside of the filling matter vessel to the probe extending within the vessel. It also must provide mechanical support for the probe. For this purpose it usually has a metallic carrier element fixedly connectable to the vessel, e.g. in a cover opening thereof and supporting a guiding element guiding the wave. To avoid a short circuit, there is an insulating element between the carrier element and the conducting element. The conducting element on the one hand connects the coaxial cable usually leading to the electronics unit and on the other hand the probe installed in the vessel.

Feedthroughs for one-conductor probes usually have a coaxial structure, i.e. the conductor element is coaxially surrounded by an insulating element and a carrier element. This principal structure can be technically realized in a number of various ways in order to satisfy certain requirements such as isolation of the vessel atmosphere, pressure resistance, reception of great pulling forces at the probe, high temperatures and resistance against aggressive vessel atmospheres. Apart from the mechanical requirements to the feedthrough, as already mentioned, the electrical requirement of guiding the wave without great impedance jumps must be observed. This requirement can be fulfilled for the coaxial conductor within the feedthrough. Examples of such approaches for electrically as well as mechanically suitable feedthroughs can be found in EP 0 773 433 Al, EP 0 780 664 A2 and WO 98/25109.

All feedthroughs described there give indications as to how the conductor impedance within the feedthrough is to be kept essentially constant. For matching the inevitable impedance jump between the coaxial feedthrough and the adjacent probe (also called single conductor) there are no solutions in these references. The impedance jump mentioned is very pronounced in the usual case and therefore poses particular interference problems. This is because the line impedance of a single conductor is in the order of 300Q. With coaxial cables, an impedance results from the ratio of the diameter of an external conductor D to an internal conductor d as well as the dielectric constant of the interposed insulating material. The greater the D/d ratio and the smaller the dielectric constant, the greater the impedance. The measure D of the external conductor is in practice limited to the top by usual vessel opening sizes, the measure d of the internal conductor is limited to the bottom by the required mechanical stability of the conductor element. Thus, in total, the line impedance is limited toward greater values by the given limitations of the mechanical dimensions.

Easily realizable impedance values for coaxially constructed feedthroughs are between 500 and 1 00 and are usually dimensioned such that they pass on the impedance of the coaxial cable connecting it to the electronics unit. This means that the impedance of the coaxial feedthrough is usually in the range of the standard values of 50M or 750. From this consideration results an impedance jump at the feedthrough/single conductor interface by considerably more than a factor 2. The previously known improvement of the line matching from the feedthrough impedance to the impedance of the single conductor is described for example in the already mentioned DE 44 04 745 C2. By an adjustment funnel adjacent to the feedthrough the wave resistance is not increased by a step change but continuously increased from the low value of the feedthrough to the high value of the single conductor. The drawback with this approach is the space requirement of the funnel within the vessel and the ID- 7 O risk of filling matter adhesions within the funnel as well as damaging effects of the vessel atmosphere on the funnel.

0 I Summary of the Invention It is therefore an object of the present invention to provide a feedthrough for radio Sfrequency signals improved with respect to interfering echoes in a TDR filling level Smeasuring device.

C, 10 According to the present invention there is provided a feedthrough for an electrical radio frequency signal in a time domain reflectometry filling level measuring device having: cI- a conductor element into which at an inlet position the electrical radio frequency signal is to be fed in and which at an outlet position transfers the electrical radio frequency signal to a probe for guiding the radio frequency signal, a mechanical carrier element, an insulation present between the mechanical carrier element and the conductor element, an attenuating element attenuating the electromagnetic wave, wherein the attenuating element is present between the carrier element and the conductor element of the outlet position and which is designed such that the impedance of the feedthrough and the impedance of the probe adjacent to the outlet position are essentially mutually matched at the outlet position.

A feedthrough according to the present invention of the kind mentioned above is characterized for the first time by the impedance of the feedthrough and the impedance of the probe serving to conduct rather than to radiate the radio frequency signal being matched at the outlet position. In contrast to the prior art, thus for the first time, the impedances at the outlet position of the feedthrough are taken into account, and by the impedance matching in this area previously arising interference reflections are largely avoided or at least reduced.

On the one hand, the new impedance matching may be achieved by increasing the impedance within the feedthrough to the higher impedance of the probe by different structural measures. Such embodiments of the invention comprise for example H,\Pcabral\Keep\speci\2002214051.2.doc 15/11/06 I\ 7a dimensional changes of a number of components within the feedthrough (cf. e.g. Figs. 2 and On the other hand, it is also possible to reduce the higher impedance of the probe

O

z to the lower impedance of the feedthrough. This is done by adding suitable components in the area of the outlet position, such as it is shown for example in Figs. 6 to 8 in various variants. Moreover, it can be seen from Fig. 9 that both approaches may also be combined with each other.

SAccording to the first approach, therefore, the impedance at the outlet position for matching to the impedance of the probe adjacent at the outlet position is considerably N 10 higher than the impedance at the inlet position. The impedance at the outlet position -therefore should be H;\Pcabal\Keep\apec1\2002214051.2.doC 15/11/06 -8matched as closely as possible to the impedance of the probe in order to avoid interfering impedance jumps having the above-mentioned drawbacks. An improvement over the prior art is achieved simply by keeping the impedance of the probe not more than 1.5 times higher than the impedance of the feedthrough at the outlet position, which according to the invention is to be understood as an essential matching of the impedances. I.e. that for example using the above-mentioned conditions of the prior art the impedance of 50 at the inlet position is increased up to an impedance at the outlet position of 200(2 so that the difference to the impedance of the probe is as little as 100, greatly reducing interfering reflections over and above the state of the art.

The invention is based on the novel principle to avoid an undesired impedance jump, unlike the prior art, not by using space occupying means downstream of the feedthrough, but by using the feedthrough itself for impedance matching at the transition (outlet position) from the feedthrough to the probe disposed outside of the feedthrough. This can be done by suitably choosing individual component materials, by forming individual or several feedthrough components in a new way or by using elements easily integrated in the feedthrough and causing impedance matching, such as a discrete resistor, a wave attenuating element or a radio frequency transformer. It goes without saying that combinations of two or more of the above approaches are also possible.

In a first embodiment of the present invention a matching between the impedance of the connection to the electronics unit and the impedance of the probe is achieved in a coaxially structured feedthrough by continuously changing the impedance of the conductor. This can be realized for example by a continuous variation of the ratio of the inner diameter of the carrier element and the diameter of the conductor element, each referring to a sectional plane normal to the wave propagation direction. For example a continuous impedance change from 500 to 3000 is thus possible. The longer the zone of the continuous impedance change can be made, the freer of reflections it will be. In an ideal case the wave is conducted without major reflections from one end of the feedthrough to the other end and does not see a major impedance change even at the interface with the probe. The diameter ratio variation realized continuously or, alternatively, in several steps, can be achieved by a tapering form of the conductor element, by a conical internal contour of the carrier element or a combination of the two possibilities.

At a required minimum diameter of the conductor element and, depending on the vessel, at a limited outer diameter of the carrier element, however, this approach is possible only under certain conditions, since in this case, impedances of arbitrary height are not possible within the feedthrough.

In a further embodiment of the present invention without the above limitation, the structure of the feedthrough is not purely coaxial, but a so-called two-wire conductor is realized within the metal carrier element. By continuously varying the thickness and the distance of the additional conductor element as well as by limiting the length of the second conductor element to the structural length of the feedthrough, a continuous impedance increase of the line may be achieved in spite of the limited cross-sectional area. In this construction, only the probe keeps protruding into the vessel having all well known advantages of this simple probe.

The impedance of a two-wire conductor can be made relatively high, such as about 250L, in particular if its structure is asymmetrical the wire cross sections are different) in a small cross-sectional area used. However, impedances of smaller than 100 Q are also easily realizable in the same limited cross-sectional area using the two-wire conductor.

In a further embodiment of the present invention the impedance matching of for example the coaxial line having a small cross-sectional area to the single conductor protruding into the vessel is improved by attenuating the wave within the feedthrough. This dissipative matching, which is well known in principle, may be realized for the radio frequency feedthrough either by inserting a discrete ohmic resistor or by using a wave attenuating material as the line dielectric material. As a preferred exemplary embodiment, a material having a fine distribution of conductive pigments within the filling material, such as for example fine graphite powder which is admixed to a Teflon mass, should be mentioned. Through the volume, the form and the conductivity of the wave attenuating material, the desired impedance matching may be optimised. In principle, the present approach is based on a first parallel connection of the impedance of the probe and the ohmic resistor or the resistor formed by the wave attenuating material. This first parallel connection in turn is connected in parallel with the impedance of the coaxial line consisting of the conductive element and the carrier element. Taken together, a parallel connection of the impedance of the probe, the discrete ohmic resistor or the wave attenuating material and the impedance of the coaxial line in the area of the outlet position is provided.

The wave attenuation within the feedthrough of course also causes an amplitude reduction in the filling matter reflection to be evaluated, but in comparison, the interfering echo at the connection position between the feedthrough and the single conductor is reduced to a greater extent so that overall a more favourable ratio between useful echo and interfering echo is achieved. This method of the dissipative matching may be applied in an advantageous manner both to the coaxial feedthroughs known from the prior art and to the above-described feedthroughs having two-wire conductors.

In a further embodiment of the present invention, the relatively low impedance of the wave guide within the feedthrough and the relatively high impedance of the single conductor probe are matched by means of a radio frequency transformer. In principle, such impedance transformation using a transformer is well known in the art. The impedance changes from the input to the output of the transformer is the square of the voltage transformation ratio or the winding ratio. However, in the case of the feedthrough to be optimised, the transformer must be located at the position of the existing impedance jump, i.e. at the transition of the feedthrough to the single conductor. This may be solved by supporting the single conductor in -11the feedthrough in isolation and to connect the transformer to the beginning of the metallic single conductor in the vicinity of the end of the feedthrough. By varying the winding ratio of the transformer, a matching of various, theoretically any, input and output impedances may be achieved.

In this embodiment of the present invention, the impedance jump, as well as the drawbacks associated with it, are so to speak "prevented" by the radio frequency transformer. In contrast to the embodiment according to Fig. 2, in which the matching of the impedances at the outlet position is achieved by essentially increasing the impedance within the feedthrough toward the outlet position, namely in the direction of the higher impedance of the probe, here the impedance within the feedthrough need not be significantly changed. At the outlet position a mutual impedance matching is now "forced" by the radio frequency transformer.

Short description of the drawings For further explanation and better understanding several embodiments of the present invention will be described and explained in the following in more detail with reference to the accompanying drawings, in which: Fig. 1 is a radio frequency feedthrough for a single conductor filling level probe according to the prior art; Fig. 2 is a first embodiment of a radio frequency feedthrough for a single conductor filling level probe according to the present invention having a coaxial conductor having a varying cross section ratio between the outer and the inner conductor; -12- Fig. 3 is a second embodiment of a radio frequency feedthrough for a single conductor filling level probe according to the present invention having a two-wire conductor limited to the feedthrough; Fig. 4 is a schematic diagram showing a dissipative matching; Fig. 5 is an exemplary graphical representation for determining a useful and an interference amplitude in a dissipative impedance matching according to the schematic diagram of Fig. 4; Fig. 6 is a third embodiment of a radio frequency feedthrough for a single conductor filling level probe according to the present invention having a dissipative impedance matching by means of a discrete resistor; Fig. 7 is a fourth embodiment of a radio frequency feedthrough for a single conductor filling level probe according to the present invention having a dissipative impedance matching by means of a wave attenuating material; Fig. 8 is a fifth embodiment of a radio frequency feedthrough for a single conductor filling level probe according to the present invention having a radio frequency transformer for frequency matching; and Fig. 9 is a sixth embodiment of a radio frequency feedthrough for a single conductor filling level probe according to the present invention in which two above-described variant approaches are used in combination.

Description of preferred embodiments of the invention 13 The radio frequency feedthrough of the prior art as shown in Fig. 1 includes a carrier element 1 which is usually made of metal and which may be screwed, via an external thread 2 in its bottom portion, into an opening 3 of the vessel cover 4 having a corresponding internal thread. For assembly, the carrier element 1 has an external hexagonal shape 5. Coaxially with the carrier element 1, the feedthrough also has a metal conductor element 6. It connects the single conductor probe 7 protruding into the vessel with the plug connector 10, which in turn is connected with the electronics unit 8 via the coaxial cable 9. The plug connector 10 has a resilient pin 10 Oa received by a bore 6a at one end of the conductor element 6 in a contacting way. The outer sheath 10b of the plug connector 10 contacts the metal disc 11 via springs (not shown in more detail), the metal disc 11 in turn being fixedly mechanically and electrically connected to the carrier element 1. By suitably choosing and dimensioning the various mechanical elements of the feedthrough it is possible to achieve that the line impedance of the coaxial line 9 is continued in a matched way from the coaxial line structure of the feedthrough consisting essentially of the carrier element 1 and the conductor element 6.

For mutual insulation and mechanical fixing, the gap between the conductor element 6 and the carrier element 1 is mostly filled by non conducting insulating elements 12 and 13. The conductor element 6, for receiving pressing and pulling forces applied to the probe 7, has an enlargement 6b having adjacent conical sections 6c and 6d. These transmit the forces arising from the vessel to the insulation elements 12 and 13, which in turn are supported by the carrier element 1 and the metal disc 11. Sealing rings 14 and 15 ensure that gases from the inside of the vessel cannot pass through the feedthrough to the outside. At its top end, the feedthrough is connected with the electronics housing 16 which is only partially shown.

In the electronics unit 8 a for example pulse-shaped transmitting signal is generated which is passed to the single conductor probe 7 via the coaxial cable 9, the plug connector 10 and the coaxial feedthrough. At the connecting point 17 between the conductor element 6 and the probe 7, the line impedance changes relatively abruptly, which is inevitable for this kind of 14construction. Assuming that the impedance of the coaxial conductor is usually 50 and the coaxial feedthrough largely maintains this impedance, there is a resulting impedance jump at the connection or outlet point 17 from about 50Q to about 300Q. Due to this, a major part of the transmitting signal is reflected back to the electronics unit 8 at this point. The remaining portion of the wave is passed along the probe to the filling matter surface. There, depending on the dielectric constant of the filling matter, a further part of the wave is reflected. The now remaining part penetrates the filling matter and is reflected at the end of the probe or partially absorbed by the attenuating characteristics of the filling matter. The back reflected echoes from the connecting point 17, the filling matter surface and perhaps the probe end are received, processed and evaluated in the electronics unit 8, as is well known from the prior art. Since the interfering echo from the connecting point 17 can be significantly stronger than the useful echo from the filling matter surface, the latter may not be unequivocally identified and evaluated, if the filling level approaches the connecting point 17.

An improvement of this minimum distance to be kept may be achieved by a feedthrough according to the invention in a first embodiment thereof, as shown in Fig. 2. In Fig. 2, components corresponding to those of Fig. 1 have been indicated by the same reference numbers. The conductor element 6 here has a continually reduced diameter d, while the internal diameter D of the carrier element 1 remains largely constant. The coaxial line thus formed can be structured in such a way that it has an essentially equal impedance to the connecting plug 10 and the feeder line 9 when connected to said connecting plug 10. This is possible through a suitable choice of the diameter ratio Dl1/dl while simultaneously observing the dielectric constants of the insulation element 13.

Near the connecting point 17, the diameter ratio D2/d2 has changed in such a way that the resulting impedance of the coaxial conductor has further approached, or is equal to, the impedance of the single conductor 7. The result of this matching is a reduction of the interfering reflection from the connecting point 17 and a simultaneous amplitude increase of the reflection from the filling matter surrounding the single conductor. This continuous impedance matching by means of a coaxial line within the feedthrough can also be replaced by a plurality of impedance step changes, if necessary. The finer the step changes the freer of reflections the matching will be. Instead of changing the diameter d of the conductor element 6, also the internal diameter D of the carrier element 1 or both may be changed correspondingly. It is also possible, while keeping the diameter ratio D/d constant, to change the dielectric constant of the insulating element 13 continuously or in step changes in such a way that the impedance of the coaxial conductor is varied.

A second exemplary embodiment of a feedthrough according to the invention is shown in Fig.

3. Herein the conductor element 6 is no longer concentrically positioned within the carrier element 1, but offset from the center, so that a second conductor element 18 may be accommodated next to it. This second metallic conductor element 18 is mechanically and electrically connected to the metal disc 11 and therefore provides a conductive connection to the external conductor of the coaxial cable 9. It has a cylindrical section 18a sealed by a sealing ring 19, and a tapering section 18b ending in a point 18c. Together with the conductor element 6 it forms a two-wire or parallel-wire conductor. The distance and the diameter of the section 18b are continuously varied with respect to the conductor element 6. The asymmetrical two-wire conductor thus formed guides the wave through the feedthrough. Its impedance is determined by the mutual distance and the diameter of the two conductors. The stronger the asymmetry and the farther the mutual distance, the greater the impedance. By appropriately choosing the distance and diameter contour of the second conductor element 18, a continuous impedance variation of the parallel-wire conductor may be realized from values smaller than 100 up to values greater than 300O. With this, the impedance jump at the connecting point 17 may be considerably reduced thus reducing the amplitude of the associated interfering echo, and generally a shorter minimum distance from the filling matter to the feedthrough is realizable.

16- In order for the end of the two-wire conductor or the end of the second conductor element 18c not to generate an additional interfering echo, it may be shorted with the carrier element 1 as shown in Fig. 2. By opposite echo polarities of the echo from this short circuit and of a possibly remaining reflected portion of the wave at the connecting point 17, which through the transition from a low impedance to a high impedance is nearer open-circuit conditions, these two portions could compensate each other and therefore further reduce the resulting interfering echo.

As an alternative to the short circuit of the end 18c, the latter may also be structured by a ohmic connection to the carrier element 1 to have electric wave absorbing properties, so that an additional reflection of the wave does not arise. An example of such an ohmic connection is a low-inductivity SMD resistor in the order of 200 to 500M.

If it has not been entirely possible to match the impedance of the portion of the parallel-wire conductor within the feedthrough facing away from the vessel by determining its geometry in such a way that it is relatively closely matched to the coaxial cable 9 and the plug connector a radio-frequency suitable transformer can be connected immediately after the plug connector 10. It transforms the signal arriving from the cable and the plug connecter via the coaxial system to the two-wire system consisting of the two conductor elements 6 and 18. A voltage transformation means at the same time an impedance transformation and therefore provides the possibility of matching different line impedances. Suitable transformers having a special structure and suitable ferrite material are commercially available. The corresponding circuitry is well known to one skilled in the art which is why it is not described here in more detail.

Another approach for improving the impedance matching between the feedthrough and the single conductor probe as mentioned above is done by a parallel connection of a dissipative or wave attenuating component. In principle, this method is well known to one skilled in the art.

-17- In Fig. 4 an example is shown where a first line having the wave resistance Z 1 is connected to a second line having in this case the higher wave resistance Z2. The impedance jump arising at the connection point may be reduced or avoided by the parallel connection of the matching resistor. For the transmission signal arriving through the first line, an impedance results at the connection point, which can be calculated from the parallel connection of the line impedance Z2 and the matching resistor. Hereby an ideal impedance matching is realizable, however associated with a loss of signal amplitude of the transmission signal forwarded on the second line. In other words, this type of matching causes a considerable reduction or a complete elimination of the interfering echo from the connection point at the expense of the amplitude of the useful echo from the filling matter surface generated on the further signal path.

Fig. 5, in a graph for the example of a first line having Zl=50 and a second line having Z2=3000, shows how the amplitudes of the useful and the interference echoes vary with different matching resistances. The amplitudes are plotted as a function of the transmission signal amplitude. The interfering echo represents the reflection from the connection point of the two lines, the useful echo in this case represents the total reflection of the wave at the end of the second line. It can be seen from the graph that as the matching resistance gets smaller, the amplitude of the interfering echo decreases more quickly than the one of the useful echo.

In the example shown the interfering echo is completely eliminated with a matching resistance of 60Q, while the useful echo still has a relative amplitude of 0.17. When there is still a sufficient transmission signal amplitude, sufficiently strong useful echoes can still be obtained while the ratio of the useful echo to the interfering echo is significantly improved.

Fig. 6 shows how the method of the dissipative matching can be applied to the feedthrough according to the prior art shown in Fig. 1. Corresponding components have again been indicated using the same reference numerals as in Fig. 1. In addition to the prior art, in the feedthrough according to Fig. 6, in the vicinity of the connection point 17, a matching resistor has been connected. This can be for example a low-inductivity SMD resistor soldered onto 18a printed circuit board, on the one hand connected with the conductor element 6 via short conduction paths, and on the other hand connected to the carrier element 1. In order to protect the resistor 20 against influences of the vessel atmosphere, it has been cast into a protective layer 21. The value of the matching resistor can be chosen according to the criteria of which ratio between the interfering echo and the useful echo is to be achieved and how big a reduction in the useful echo amplitude can be afforded at the same time. The principal facts have already been described in connection with the graph of Fig. 5. Instead of the short conductors, advantageously also flexible contacting between the printed circuit board and the carrier element or conductive element is possible. To do this, the printed circuit board is for example formed to be annular and concentric with the conductor element. At both its outer and inner edges it has a conduction path also in annular form. The soldered-on SMD resistor is connected to the outer annular conduction path with its one contact, and with the inner annular conduction path through its other contact. Contacting is done for example using a first annular spiral spring contacting the outer conduction path and at the same time touching the inner wall of the carrier element all the way round, and by a second annular spiral spring contacting the inner conduction path and at the same time contacting the outer wall of the conductor element all the way round. This type of contacting allows for a certain amount of mutual displacement of the contacting parts, such as by temperature expansion of the materials. Instead of a single SMD resistor, two or more resistors may be connected in series or parallel in order to realize on the one hand any required resistance values outside the usual standard values and on the other hand to reduce feeder line inductivity through a parallel connection. Furthermore, at this point, with respect to temperature resistance, a circuit structure on the basis of a ceramic or PTFE substrate would suggest itself. Instead of the discrete resistors, also the circuit elements may be applied on the substrate using thick film or thin film technology.

In Fig. 7, an alternative form of the dissipative matching of the feedthrough is shown. Instead of the resistor of Fig. 6, a wave attenuating layer 22 is provided directly adjacent to the -19connecting point 17. As examples of materials from which such a layer can be made, a distribution of conductive pigments in Teflon or a distribution of graphite powder in Teflon can be mentioned. By appropriately choosing the form, thickness and material properties of the wave attenuating element 22, analogous to the matching resistance value, the most advantageous tradeoff between eliminating the interfering echo and attenuating the useful echo may be found. By a tapering form of the element 22 for example, the interface with the insulating element 13 may be formed in such a way that a smooth transition is created for the wave propagation without additional reflections.

A further exemplary embodiment of impedance matching using a radio frequency transformer is shown in Fig. 8. Again, the conductor element 6 together with the carrier element 1 forms a coaxial line for example passing on the impedance of cable 9 and plug connector 10 without reflections. The single conductor 7 only protrudes a little over the connecting point 17 into the feedthrough. The conductor element 6 and the single conductor 7 are not directly connected one to the other. Rather, an insulating support element 23, such as consisting of a disc 23a and a stud 23b, is inserted between the two conductive elements 6 and 7. The stud 23b having an external thread is for example screwed into an internal thread of the single conductor 7. The support element 23 is supported by for example a metallic disc 24, transferring the pulling forces acting on the single conductor 7 to the carrier element 1. Between the end of the coaxial conductor formed by the conductor element 6 and the carrier element 1 and the beginning of the single conductor 7, a radio frequency transformer 25 is electrically connected. Its winding ratio is adapted in such a way that it transforms the impedance of the single conductor 7 to the impedance of the coaxial line within the feedthrough, i.e. it transforms e.g. 300 to 50Q, which would correspond to a winding ratio of 1 2.45. In order to protect the transformer 25 against influences of the vessel atmosphere it is embedded in a cast material 26.

Finally, Fig. 9 shows an example of a combination of two above-described principal approaches for improving impedance matching. The feedthrough construction having an internal parallel-wire conductor according to Fig. 3 now also has an additional element 27 consisting of wave attenuating material and, in analogy to the disc 22 of Fig. 7, has an impedance matching effect while at the same time reducing the amplitude of the useful signal.

By varying both the geometry of the two-wire conductor and the geometry and material properties of the element 27, the effects of the two matching measures are to be mutually adjusted in a manner obvious to one skilled in the art, so that the result comes as close as possible to the ideal of a minimal interfering echo and a strong useful echo.

Further combinations of different elements from the above-described matching methods are also possible and evident to one skilled in the art without being exhaustively mentioned here.

For example, the two-wire approach of Fig. 3 may be combined with the resistance matching from Fig. 6 or the embodiment having a transformer as mentioned in connection with the twowire approach may also be applied to an approach using dissipative matching.

In the claims which follow and in the preceding description of the invention, except where the context requires otherwise due to express language or necessary implication, the word "comprise" or variations such as "comprises" or "comprising" is used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.

Claims

2. The feedthrough according to claim 1, characterized in that the feedthrough includes a discrete ohmic resistor which enables the resulting impedance from the parallel connection of the impedance of the probe and the resistor at the outlet position is essentially matched to the impedance of the coaxial line formed by the conductor element and the carrier element.
3. The feedthrough according to claim 2, characterized in that the ohmic resistor is located at the outlet position between the carrier element and the conductor element.
4. The feedthrough according to claim 2 or 3, characterized in that the ohmic resistor is embedded in a cast material. The feedthrough according to claim 1, characterized in that the attenuating element is a wave attenuating material that is arranged around the outlet position within the feedthrough.
6. The feedthrough according to claim 5, characterized in that the wave attenuating element is of a material having a mixture of conductive pigments embedded in an insulating filler material.
7. The feedthrough according to claim 5 or 6, characterized in that the wave H:\Pcabral\Keep\apeci\2002214051.2.doc 15/11/06 22 attenuating element at least partially fills the gap between the conductor element and carrier element.
8. The feedthrough according to any one of claims 5 to 7, characterized in that the wave attenuating element has an essentially cylindrical shape.
9. The feedthrough according to claim 5 to 7, characterized in that the wave attenuating element has an essentially tapering shape.
10. The feedthrough according to any one of the preceding claims, characterized in that, in addition, a radio frequency transformer is present which, in the area of the outlet position, electrically connects the carrier element to the conductor element and essentially matches the impedance of one to the other.
11. The feedthrough according to claim 10, characterized in that the radio frequency transformer is embedded in insulating and protective cast material.
12. The feedthrough according to claim 10 or 11, characterized in that the radio frequency transformer at least partially transforms the impedance of the conductor within the feedthrough to the impedance of the probe.
13. The feedthrough according to any one of the preceding claims, characterized in that at least an additional conductor element is present which is limited to the area of the feedthrough.
14. The feedthrough according to claim 13, characterized in that the additional conductor element is electrically coupled to the carrier element. The feedthrough according to claim 13 or 14, characterized in that the distance and the diameter ratio of the two conductor elements changes at least over a partial length of the feedthrough.
16. The feedthrough according to claim 15, characterized in that the distance and the diameter ratio of the two conductor elements continuously changes at least over a partial length of the feedthrough.
17. The feedthrough according to any one of claims 14 to 16, characterized in H,\Pcabral\Keep\speci\2002214051.doc 03/05/06 23 that the additional conductor element is connected at its end to the carrier element.
18. The feedthrough according to any one of claims 14 to 17, characterized in that the additional conductor element is connected at its end to the carrier element via a resistor.
19. The feedthrough according to claim 1, characterized in that the impedance at the outlet position to be matched to the impedance of the probe adjacent to the outlet position is considerably higher than the impedance at the inlet position. The feedthrough according to any one of the preceding claims, characterized in that the conductor element is arranged to be coaxial with the carrier element.
21. The feedthrough according to claim 1, characterized in that the impedance changes continually from the inlet position to the outlet position.
22. The feedthrough according to claim 1, characterized in that the impedance changes in steps from the inlet position to the outlet position.
23. The feedthrough according to claim 1, characterized in that the internal diameter of the carrier element changes from the inlet position to the outlet position.
24. The feedthrough according to claim 1, characterized in that the external diameter of the conductor element is reduced from the inlet position to the outlet position. The feedthrough according to claim 1, characterized in that the conductor element has a tapering form at least in portions thereof.
26. The feedthrough according to claim 1, characterized in that the carrier element has a conical inner contour at least in portions thereof.
27. The feedthrough according to any one of the preceding claims, characterized in that the dielectric constant of the insulation changes from the inlet position to the outlet position. H.\Pcabral\Keep\speci\2002214051.doc 03/05/06 24
28. A filling level measuring apparatus for measuring the filling level using echo time measurement of a guided electromagnetic wave, comprising an electronics unit for generating the electromagnetic wave and for evaluating received echo signals, and a radio frequency feedthrough coupled with the electronics unit, according to any one of the preceding claims. Dated this 3rd day of May 2006 VEGA GRIESHABER KG By their Patent Attorneys GRIFFITH HACK Fellows Institute of Patent and Trade Mark Attorneys of Australia H:\Pcabral\Keep\speci\2002214O51.doc 03/05/06